Test results from RDDs will always vary. There are too many uncontrollable variables. I've gone out with the same RDD on successive days and recorded different results.
If you're who I think you are, the Spectre you purchased from Stealth MicroSystems is a hodgepodge of parts from different generations of Spectre, most of them from the Mk IV. It can be expected to perform differently from the production Mk IV and Mk IV+ that I tested at the manufacturer's request, both of which were built some time after yours was and which had different software.
If you can describe in detail your testing procedures, the vehicles used, RDD mounting position and angle, radar detector mounting position, detector vehicle positioning and travel direction relative to the RDD-equipped vehicle's; traffic conditions--in volume and vehicle types; roadway description and topography plus the presence or absence of roadside structures, I'll be in a position to offer an opinion on why the test results vary. All of these factors play a major influence on an RDD's detection range.